Geography Reference
In-Depth Information
stations, where it is preferable to have minimal variation
in width in response to fluctuations in discharge. Space-
based estimates of this sort are particularly valuable in
areas where gauging stations are being discontinued. Cor-
relations established in this manner can be extended to
other rivers of similar morphology. Alternatively, flood
volume and depth can be derived by coupling flood
inundation maps with digital elevation models, which
can be coupled with hydrologic models to estimate dis-
charge (Smith, 1997). In a related application, remote
sensing-based measurements of flood extent can be used
to calibrate and validate numeric models of flooding
(Bates, 2004).
Aerial flights and satellite imagery will doubtlessly
remain useful remote sensing tools for flow monitor-
ing. Cloud and tree cover are major obstacles to these
approaches, however, because the sensor cannot see
the water surface. Researchers and monitoring programs
therefore are increasingly turning to active radar imagers,
which penetrate forest and cloud cover and allow water to
be distinguished from other features as well as providing
surface elevation data for broad areas (Alsdorf et al., 2007;
Schumann et al., 2009).
for each biotype (riffle, pool, glide, etc.) to derive depths
that were specific to different surface turbulence regimes
within the stream. The major limitation specific to the
correlation approaches is that water depths must be
measured in the field at the same time as imagery is col-
lected to avoid variations in discharge and channel shape
that could modify the relationship between pixel values
and depth.
Physically-based models provide an alternative to sim-
ple correlation approaches. These models are based on
the physics of how light moves through water, as is sum-
marised in Chapter 3 and reviewed by Legleiter et al.
(2004; 2009). From a management perspective, the most
readily applicable of these models are those that avoid
the need for field teams to collect field data at the time-
of-flight, which removes a major logistical constraint.
Fonstad and Marcus (2005) developed a hydraulically
assisted bathymetry (HAB) model that couples equations
describing light attenuation by the water column and the
hydraulics of open channel flow with data on discharge,
slope and channel width to map depths throughout a
stream (Figure 2.2). Their physical modeling technique
does not require field crews or data collection specific to
the project. The discharge data can come from nearby
gauging stations, slopes can come from maps or other
sensors, and width can be measured from the imagery.
Moreover, the model is sufficiently simple that the math-
ematical formulae used to compute depth estimates can
be implemented in a spreadsheet. Because the HAB tech-
nique does not require field measurements, it can be
used with historical imagery so long as discharge data are
available for a nearby site.
Some researchers have expressed concern, however,
that simple models like HAB do not consider the effects of
variables like turbidity, substrate size and colour, surface
turbulence, algae on rocks, and other factors that could
potentially complicate depth mapping. In an extensive
theoretical and empirical experiment, however, Legleiter
et al. (2009) demonstrated that these other factors can
be accounted for and accurate depth estimates achieved
by using the natural log of an appropriate band ratio
(Table 2.1). A band ratio is simply one band divided by
another, and typically the green band value is divided
by the red band value for the same pixel. Differences in
pixel values due to sediment colour, sediment size, veg-
etation, and other substrate characteristics are, in effect,
normalized by band ratios, leaving the depth signal as the
primary factor driving variations in pixel values. Legleiter
et al. (2009) also demonstrated that depth maps derived
from such ratios show remarkable resiliency across a range
2.4 Water depth
The importance of water depth to monitoring, mapping
and modeling river habitats has generated consider-
able research interest in measuring depth from optical
imagery. As long as the water is clear enough to see to the
bottom, there are three general approaches that provide
relatively accurate depth estimates (Table 2.1).
The easiest depth mapping technique is the correla-
tion approach, where the brightness of the image (i.e.,
the pixel value) at a number of locations is correlated
to field measurements of depth at the same locations.
The regression equation derived from this correlation is
then applied to the remainder of the image to estimate
water depths. There are a number of variations to the
correlation approach. The highest accuracies (Table 2.1)
are achieved by using multiple regressions where more
than one image band (e.g. red, green and blue bands)
are correlated to depth (e.g., Gilvear et al., 2007; Lejot
et al., 2007). Marcus et al. (2003) achieved high accura-
cies using 128-band hyperspectral imagery that covered
the visible and shortwave infrared wavelengths. They first
ran a principal components analysis on the water portion
of the image to remove noise and reduce the dimension-
ality of the spectral signal, then ran a step-wise regression
Search WWH ::
Custom Search